APPLICATIONS AND CASE STUDIES Wear characterization by fractal mathematics for quality improvement of machine S. Ghosh, B. Sarkar and J. Saha Department of Production Engineering, Jadavpur University, Kolkata, India Abstract Purpose – The objective of the present work is to find an alternative approach for gearbox condition monitoring using wear particle characterization incorporated with image vision systems. Design/methodology/approach – It is a quite well-known phenomenon that wear generates whenever two metallic bodies have contact with each other; hence the present work tries to investigate the effect of improper lubrication in the gearbox due to wear particle generation between gear wheels. Since the identification of wear for machine condition monitoring needs much expertise knowledge and is time-consuming using the conventional process, fractal mathematics with image morphological analysis has been utilized to overcome this situation in the present work. Findings – The type of wear has been found for the present method by utilizing the lubricant used in the system ferrographically and a great deal of image processing has been done to characterize the type of particle so that the proper maintenance strategy can be undertaken. Originality/value – Wear particle characterization is a quite common method in maintenance engineering, especially when fault diagnosis of any equipment is concerned. In the present work, the CCD acquisition of the images has been done for different particles, but one analysis amongst them has been shown in this paper. Among all other methodologies, the new technique of fractal mathematics has been used in the present work to minimize the imaging hazards and to make the system more user-friendly. Keywords Mechanical components, Image processing, Condition monitoring, Wear Paper type Research paper 1. Introduction The condition monitoring of any rotating machinery through analysis of wear debris generated in a lubricating system can assess the operational effectiveness and it can be applied to any diagnostic technology. The reliability of a particular lubricating system is directly related to the presence of the solid particulate matters contained in lubricants. Generally 50 percent failures of any equipment can occur due to contamination in lubrication system. So the detection and analysis of contamination deals with wear debris analysis both quantitatively and qualitatively. The technical judgments of wear based on the morphological analysis of the wear particles extracted from respective debris of the used lubricating oil. Owing to this reason, various techniques being adopted for identifying the wear particles for monitoring the machine condition according to their rate of production. Determination of wear particle features needs some computer image processing for improving its quality. After that features of The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at www.emeraldinsight.com/researchregister www.emeraldinsight.com/1355-2511.htm The authors wish to thank the Department of Sciences and Technology under Ministry of Science and Technology, Govt of India for their valuable support for carrying out the project. JQME 11,4 318 Journal of Quality in Maintenance Engineering Vol. 11 No. 4, 2005 pp. 318-332 q Emerald Group Publishing Limited 1355-2511 DOI 10.1108/13552510510626954